The Agent Skills Directory

[SAFE]: The skill consists entirely of high-quality instructions and frameworks for content strategy and does not include any executable code or scripts.
[DATA_EXPOSURE]: The skill references reading a specific context file (.claude/product-marketing-context.md). This is a documented practice for providing business context to the agent and does not constitute unauthorized data access.
[PROMPT_INJECTION]: No attempts to bypass safety filters, override system instructions, or extract system prompts were detected. The use of a professional persona is consistent with the skill's purpose.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external data (transcripts, survey results, and web search content), it lacks dangerous capabilities like arbitrary code execution or file writing, which significantly mitigates the risk of exploitation via untrusted data.
Ingestion points: Reads .claude/product-marketing-context.md and processes user-provided CSVs, transcripts, and web search results.
Boundary markers: Not explicitly defined for external data input.
Capability inventory: Limited to generating text-based strategy documents and tables.
Sanitization: Relies on the underlying LLM's inherent safety guardrails.

content-strategy